4618
EMILJ. MORICONI, BOHDAN KAKOCZYAND WILLIAMF. O'CONNOR
Vol. 83
excess of hydrogen at 50 lb. per square inch for four hours. anol and the hazy solution filtered. The pH of the clear T h e crude reaction mixture was filtered through Celite t o filtrate was adjusted to 9.0 by the addition of triethylamine remove the majority of the catalyst. T h e filtrate was con- and then brought down to p H 1.5 by the addition of concentrated in vacuo and the sirupy residue taken up in 200 ml. centrated sulfuric acid. Upon stirring and scratching, bright of distilled water and freeze-dried. yellow needles deposited. The yield of 6-deoxy-6-demethylThe crude reaction mixture was filtered through Celite to chlorotetracycline (31) was 1.8 g. remove the majority of the catalyst. The filtrate was conAnal. Calcd. for CPIH2107C1N2 H&O1 H20: C, 41.6; centrated in vacuo and the sirupy residue taken up in 200 ml. H. 4.46; N,4.96; C1,6.28. Found: C,44.60; FI,4.45; S , of distilled water and freeze-dried. 4 75; C1, 6.51. The freeze-dried material from five such runs was coniDedimethylamino-6-demethyl-6,~2a-dideoxychforotetrabined and dissolved in 450 nil. of butanol saturated with water. The pH was adjusted to 2.0 and the solution filtered. cyclhe (17).--Zinc dust (500 mg.) was added to a solution of 6-deoxy-6-demethylchlorotetracycline (400 mg.) and anhyThe filtrate was chromatographed on a 6-inch column packed with 3.4 kg. of acid-washed Celite using a 8070 B ~ 0 H - 2 0 7 ~ drous sodium acetate (65.5 mg.) in glacial acetic acid (20 ml.) CHCla-PH 2.0 developer. A total of twelve 250-ml. cuts and water ( 9 ml.). The mixture was stirred under nitrogen were taken. The composition of the cuts was determined by for 1.5 hours and an additional 500 mg. of zinc dust was paper chromatography using a butanol-phosphate-pH 2 .O added. After 6 hours the reaction mixture was diluted with 300 ml. of ethyl acetate, water (2.1) and the undissolved system. zinc filtered off. The aqueous layer was separated and exCuts 1 and 2 were essentially pure 6-deoxy-6-demethyltracted thrice with ethyl acetate. The combined ethyl acechlorotetracycline. Cuts 3 through 9 were a mixture of tate layers were washed with 1 N sulfuric acid and thrice 6-deox~--A-clemethylchlorotetracycline, 6-demethyl-6-deoxytetracycline and a small amount of starting material. Cuts with water. The organic layer was separated, dried over anhydrous magnesium sulfate and concentrated t o dryness 3 through 9 were combined and concentrated to 400 ml. The PH was adjusted t o 2.0 and the solution filtered. T h e zn vacuo. The residual orange solid was slurried in 10 ml. of clear filtrate was columned as before. Paper strips indicated toluene and again concentrated t o dryness in vacuo. The that cuts 1 through 7 mvere essentially pure 6-deoxy-6-de- yield of crude dedimethylamino-6-demethyl-6,12a-dideoxychlorotetracycline (17) was 97 mg. An analytical sample was inethylchlorotetracycline . washing with a small volume of chloroform. The Cuts 1 and 2 from the first column and cuts 1 through 7 obtained by spectrum, the ultraviolet spectrum A(: NaoH"aH) from the second column were concentrated t o dryness in infrared 495, 469, 375, 262 mp) and the chromatographic behaviacuo. The residue was taken up in 300 ml. of water the p H adjusted t o 2.0 with hydrochloric acid and filtered. The ior (Ri0.61 in a butyl acetate-formamide-acetone system u b filtrate was freeze-dried t o yield 4.85 g. of 6-deoxy-6-de- ing paper buffered a t PH 6.0) of this material were identical t o the corresponding properties of the synthetic amide 17. niethylchlorotetracycline hydrochloride. Further purification was accomplished as follows: 2.3 g. Anal. Calcd. for C,oHl&CIOa: C, 58.6; H, 4.14; N, of the freeze-dried material was dissolved in 55 ml. of meth3.60. Found: C, 58.69; H, 4.64; N,3.88.
[CONTRIBUTION FROM
THE
DEPARTXENT O F CHEMISTRY, FORDHAM UNIVERSITY, NEW Y o u 58, N. Y.]
Ozonolysis of Polycyclic Aromatics. VIII.'
Benzo [a]pyrene2~3
BY EMILJ. MORICONI, BOHDAN RAKOCZY AND WILLIAMF. O'CONNOR RECEIVED J U N E 30, 1961 Ozonization of benzo[a]pyrene ( I ) in methylene chloride and 3 : 1 methylene chloride-methanol with o r e molar equivalent of ozone produced a mixture of quinones from which could be isolated benzo[a]pyrene-3,6-dione(11) and benzo[a]pyrene1,6-dione (111) in 1: 3 ratio, and trace amounts of benzo[a]-pyrene-4,5-dione. KO solvent effects were obrerved. Ozonolysis of I with two niolar equivalents of ozone gave 7H-benz[de]anthracen-7-one-3,4-dicarboxylic acid ( I V ) while four, six and eight molar equivalents of ozone produced only 12-anthraquinonedicarboxylic acid (VII). I n all cases unreacted I was recovered. Ozonolysis of the quinone mixture also produced I V and VII. The stoichiometry of these various ozonizations is unclear but does suggest predominant attack a t positions 1, 3 and 6 in I by more than one molecule of ozone. Several inechanisms are proposed. There now does not seem to be any simple correlation between K- and L- region additivity toward ozone and carcinogenicity in the series anthracene, benzlajanthracene, dibenz[a.h]anthracene and I.
Introduction Since our original investigation of the ozonolysis of ~ h e n a n t h r e n e , ~we " . ~have ~ actively investigated the reaction between ozone and polycyclic aromatic hydrocarbons. Our purpose was twofold: (i) to ehcidate the nature of ozone attack on aromatic systems; and (ii) to correlale the relative carcino(1) P a r t V I I , E. J. SIoriconi, W. F. O'Connor, W. J. Schmitt, G. \%'. Copswell and B. P. P J r e r , J . A m . C h o n . Soc., 82, 3441 (19GO) ( 2 ) Presented in part a t the Symposium on Ozone Chemistry, 135th National Meeting of the American Chemical Society, Atlantic City, N. J . , Sept., 1959. ( 8 ) This research was supported by a grant C-3325(C3) from t h e V , S. Public Health Service, National Cancer Institute. (4) (a) W. J. Schmitt, E J. Moriconi and W. F. O'Connor, J . A m . Cheiir. S i x . , 7 7 , 6640 (1956); ( b j W. F. O'Connor. W. J. Schmitt and t; J . Sluriwni, Znd. E u g . Chenz., 49, 1701 (1957); (c) E. J . illoriconi. \\', F.O'Conuor and I,. B. Taranko, Arch. Biockem. and Biophys., 8 3 , 2 3 3 (10.59): (d) E. J. Moriconi, W. F. O'Connor and F. T . Wallenberger, Chrmisfvy a n d Zndrrslvy, 22 (1959); (e) E. J . Moriconi, W. F. O'Connor and F. T . R'allenberger, J . A m Chem. S o c . , 81, 61GB (18393; ( f ! E. J Moriconi, G. W. Cogsweli, W. J. Schmitt and W. I: O'Coniior. (,'hemistry @ I i t d u s f r y , 1591 (1Y58).
genicity of these polycyclics with the course and mechanism of the ozonolysis reaction. Published results of our studies on n a p h t h a ~ e n e , ~benz ' [a]and dibenz [ a , h ] a n t h r a ~ e n eha\-e ~-~~ led us to note a simple correlation between carcinogenicity and K- and L-region activity toward ozone in this limited ~ e r i e s . In ~ ~this . ~ paper ~ we report on the ozonization of the strong carcinogen (+ +)5 benzo [alpyrene (I).'j
++
Results Ozonization of I in methylene chloride, and methylene chloride-methanol (3 : 1) with one niolar equivalent of ozone (3.5 vol. %) at -70°, and a t 20°, gave a mixture of benzo [a]pyrene-3,G-dione (11),benzo [a]pyrene-1,6-dione (111) and benzo [a]( 5 ) G. >I Badger, Brit. J . Cahcer, 2 , 309 (194s). (li)J,. F . Fieser and F C. Sovelld [ J . A m . Chcin. Sor.. 62, is55 (1!1101] have pre\,iousiy uzonized I hut t h e material appeared either t u remain unchanged ur t o be attacked extensively, and no satisfacl
